Facial-Attribute-Detection: Optimized for Qualcomm Devices

Detects attributes (eye closeness, mask presence, eyeglasses presence, sunglasses presence) that apply to a given face. This model's architecture was developed by Qualcomm. The model was trained by Qualcomm on a proprietary dataset of faces, but can be used on any image.

This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime Precision Chipset SDK Versions Download
ONNX float Universal QAIRT 2.50, ONNX Runtime 1.27.1 Download
ONNX w8a8 Universal QAIRT 2.50, ONNX Runtime 1.27.1 Download
QNN_DLC float Universal QAIRT 2.50 Download
QNN_DLC w8a8 Universal QAIRT 2.50 Download
TFLITE float Universal QAIRT 2.50 Download
TFLITE w8a8 Universal QAIRT 2.50 Download

For more device-specific assets and performance metrics, visit Facial-Attribute-Detection on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

  • Custom weights (e.g., fine-tuned checkpoints)
  • Custom input shapes
  • Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for Facial-Attribute-Detection on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

  • Input resolution: 128x128
  • Model checkpoint: multitask_FR_state_dict.pt
  • Model size (float): 46.3 MB
  • Model size (w8a8): 12.3 MB
  • Number of parameters: 12.1M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
Facial-Attribute-Detection ONNX float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.448 ms 0 - 59 MB NPU
Facial-Attribute-Detection ONNX float Snapdragon® 8 Elite For Galaxy Mobile 0.502 ms 0 - 57 MB NPU
Facial-Attribute-Detection ONNX float Snapdragon® X2 Elite 0.412 ms 1 - 1 MB NPU
Facial-Attribute-Detection ONNX float Snapdragon® X Elite 0.851 ms 23 - 23 MB NPU
Facial-Attribute-Detection ONNX float Snapdragon® 8 Gen 3 Mobile 0.625 ms 0 - 84 MB NPU
Facial-Attribute-Detection ONNX float Snapdragon® 8 Gen 1 Mobile 1.167 ms 0 - 81 MB NPU
Facial-Attribute-Detection ONNX float Qualcomm® Dragonwing™ IQ-8275 1.32 ms 0 - 4 MB NPU
Facial-Attribute-Detection ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.866 ms 0 - 26 MB NPU
Facial-Attribute-Detection ONNX float Qualcomm® QCS8450 1.167 ms 0 - 81 MB NPU
Facial-Attribute-Detection ONNX float Qualcomm® Dragonwing™ IQ-9075 1.225 ms 0 - 3 MB NPU
Facial-Attribute-Detection ONNX float Qualcomm® Dragonwing™ IQ-X7181 0.851 ms 23 - 23 MB NPU
Facial-Attribute-Detection ONNX float Qualcomm® Dragonwing™ Q-8750 0.502 ms 0 - 57 MB NPU
Facial-Attribute-Detection ONNX w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.225 ms 0 - 58 MB NPU
Facial-Attribute-Detection ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.246 ms 0 - 63 MB NPU
Facial-Attribute-Detection ONNX w8a8 Snapdragon® X2 Elite 0.18 ms 1 - 1 MB NPU
Facial-Attribute-Detection ONNX w8a8 Snapdragon® X Elite 0.385 ms 12 - 12 MB NPU
Facial-Attribute-Detection ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 0.289 ms 0 - 82 MB NPU
Facial-Attribute-Detection ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 0.531 ms 0 - 87 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 1.364 ms 0 - 3 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 0.522 ms 0 - 4 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.398 ms 0 - 20 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® QCS8450 0.531 ms 0 - 87 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 0.501 ms 0 - 3 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 0.385 ms 12 - 12 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 2.917 ms 0 - 63 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 0.507 ms 0 - 58 MB NPU
Facial-Attribute-Detection ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 0.246 ms 0 - 63 MB NPU
Facial-Attribute-Detection ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 0.507 ms 0 - 58 MB NPU
Facial-Attribute-Detection QNN_DLC float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.426 ms 0 - 57 MB NPU
Facial-Attribute-Detection QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 0.501 ms 0 - 60 MB NPU
Facial-Attribute-Detection QNN_DLC float Snapdragon® X2 Elite 0.575 ms 0 - 0 MB NPU
Facial-Attribute-Detection QNN_DLC float Snapdragon® X Elite 0.985 ms 0 - 0 MB NPU
Facial-Attribute-Detection QNN_DLC float Snapdragon® 8 Gen 3 Mobile 0.638 ms 0 - 81 MB NPU
Facial-Attribute-Detection QNN_DLC float Snapdragon® 8 Gen 1 Mobile 1.242 ms 0 - 80 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 1.304 ms 0 - 3 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.853 ms 0 - 2 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® SA8775P 1.333 ms 0 - 60 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® SA8650P 1.333 ms 0 - 60 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® SA8255P 1.333 ms 0 - 60 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® QCS8450 1.242 ms 0 - 80 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.213 ms 0 - 2 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 0.985 ms 0 - 0 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.501 ms 0 - 60 MB NPU
Facial-Attribute-Detection QNN_DLC float Qualcomm® SA8295P 1.433 ms 0 - 51 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.204 ms 0 - 55 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.232 ms 0 - 56 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Snapdragon® X2 Elite 0.291 ms 0 - 0 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Snapdragon® X Elite 0.5 ms 0 - 0 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.272 ms 0 - 67 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 0.511 ms 0 - 73 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 1.303 ms 0 - 2 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 0.485 ms 0 - 3 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.372 ms 0 - 1 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® SA8775P 0.538 ms 0 - 56 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® SA8650P 0.538 ms 0 - 56 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® SA8255P 0.538 ms 0 - 56 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® QCS8450 0.511 ms 0 - 73 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 0.446 ms 0 - 2 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 0.5 ms 0 - 0 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 2.918 ms 0 - 53 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 0.485 ms 0 - 51 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 0.232 ms 0 - 56 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Qualcomm® SA8295P 0.694 ms 0 - 52 MB NPU
Facial-Attribute-Detection QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 0.485 ms 0 - 51 MB NPU
Facial-Attribute-Detection TFLITE float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.421 ms 0 - 78 MB NPU
Facial-Attribute-Detection TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 0.485 ms 0 - 67 MB NPU
Facial-Attribute-Detection TFLITE float Snapdragon® 8 Gen 3 Mobile 0.613 ms 0 - 92 MB NPU
Facial-Attribute-Detection TFLITE float Snapdragon® 8 Gen 1 Mobile 1.201 ms 0 - 95 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® Dragonwing™ IQ-8275 1.289 ms 0 - 26 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.811 ms 0 - 31 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® SA8775P 1.341 ms 0 - 73 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® SA8650P 1.341 ms 0 - 73 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® SA8255P 1.341 ms 0 - 73 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® QCS8450 1.201 ms 0 - 95 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® Dragonwing™ IQ-9075 1.209 ms 0 - 25 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® Dragonwing™ Q-8750 0.485 ms 0 - 67 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® SA7255P 4.19 ms 0 - 71 MB NPU
Facial-Attribute-Detection TFLITE float Qualcomm® SA8295P 1.404 ms 0 - 64 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.209 ms 0 - 47 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.239 ms 0 - 46 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 0.277 ms 0 - 67 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 0.502 ms 0 - 74 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 1.186 ms 0 - 14 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 0.452 ms 0 - 14 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.373 ms 0 - 16 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® SA8775P 0.575 ms 0 - 54 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® SA8650P 0.575 ms 0 - 54 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® SA8255P 0.575 ms 0 - 54 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® QCS8450 0.502 ms 0 - 74 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 0.476 ms 0 - 14 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 2.906 ms 0 - 47 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 0.484 ms 0 - 45 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 0.239 ms 0 - 46 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® SA7255P 1.05 ms 0 - 49 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Qualcomm® SA8295P 0.698 ms 0 - 47 MB NPU
Facial-Attribute-Detection TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 0.484 ms 0 - 45 MB NPU

License

  • The license for the original implementation of Facial-Attribute-Detection can be found here.

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support